Contact
Products
Products
Collaborative robot arm assembling a circuit board alongside a human worker

Collaborative robots

Collaborative robots, or cobots, are designed to operate in environments where humans and automated systems work together. Unlike traditional industrial robots operating within highly controlled spaces, collaborative robotic systems can benefit from advanced perception technologies that help them interpret their surroundings and respond to dynamic environments.

Cameras, 3D sensing, radar, and edge processing can provide the environmental information required for advanced robotic perception. By processing sensor information close to the source, intelligent sensing architectures can support machine vision, human detection, object awareness, and real-time robotic automation. Vision-processing technologies used for robotics already support machine vision, multi-sensor architectures, 3D perception, and 360-degree vision.

Enabling Key Capabilities

Cobots operate in environments that may include people, machinery, tools, and moving objects. 

Visual and other sensing modalities provide environmental information that can be used by robotic perception systems to identify and interpret what is happening around the machine.

Machine vision enables robotic systems to capture and process visual information for automated tasks. High-performance image-processing architectures can support multiple high-resolution sensors while preparing camera information for computer-vision algorithms. Machine vision is explicitly identified among current robotics and industrial vision-processing applications.

Cameras and radar can provide complementary approaches to object awareness.

Radar is used for object detection in robotic automation and provides real-time detection and distance measurement capabilities.

Multi-camera architectures can provide broader visual coverage around a robotic platform. Vision-processing technologies already support 360-degree vision and multiple high-resolution camera inputs.

Advanced Image Signal Processing

A camera does not directly provide optimized information to an AI system. Raw image data first needs to be processed. An Image Signal Processor, or ISP, performs this critical function between the camera sensors and perception algorithms.

Advanced ISP architectures can support:

Multiple simultaneous camera inputs

High-dynamic-range imaging

Low-light optimization

Color processing

Image enhancement

Multiple high-resolution sensors

Low-latency image processing

RGB and infrared sensor configurations

For cobots, this creates a streamlined perception pipeline:

Sense. Process. Understand. Interact.

Building a Perception Architecture for Collaborative Robotics

Physical AI connects digital intelligence to systems that interact with the physical world.

For cobots, perception represents an important part of that connection.

A scalable architecture can combine:

Multi-Modal Sensors

Cameras

Depth

IR

Thermal

Radar

Edge Sensor Processing

ISP

Geometric Processing

Signal Processing

Edge AI

Computer Vision

Neural Processing

DSP

Environmental Perception

Robotic System

The objective is not simply to collect more sensor information, but to efficiently transform that information into a representation of the physical environment that downstream robotic systems can use.

Products

iND880 camera video processor chip in a BGA package, top and underside

Camera Video Processor

TR2_060_095 60 GHz radar transceiver chip (T2R95)

60 GHz Radar Transceiver

Autonomous forklift in a warehouse scanning its path with perception sensors

Perception Software

Resources

indie Semiconductor Introduces Class-leading Computer Vision Processor Family

Explore advanced image-processing capabilities including multi-camera operation, HDR, low-light processing, and proprietary ISP technology.

indie Semiconductor’s Latest Vision Processor Design-win Momentum Accelerates, Awarded Sensor Innovation of the Year

Learn more about high-dynamic-range vision, low-light capabilities, real-time processing, and memory-efficient vision-system architectures.